Since the discovery of high electrical conductivity in "doped" polyacetylene films in the mid-1970's, the field of electroactive polymers has undergone explosive growth. The great interest in these materials stems from their potential use in electronic and optical applications. Electrical conductivity is typically achieved via oxidative (or, more rarely, reductive) doping of the neutral polymers, a practice which is often accompanied by reduced processibility and environmental stability. Hence a major goal in this field is the design and synthesis of processible polymers with low or zero bandgaps.
The potential benefits from such low gap polymers are well recognized and recent theoretical approaches have focused on bond length alternation (Bredas, et al., 1986; Toussaint, et al., 1989-2; Toussaint, et al., 1989-1; Bredas, J. L., 1985; Bakhashi, et al., 1987; Bredas, J. L., 1987; Kertesz, et al., 1987; Hanack, et al., 1991) and variations in occupancy of frontier orbitals (Tanaka, et al., 1985; Tanaka, et al., 1987; Tanaka, et al., 1989; Tanaka, et al., 1988) to identify likely low E.sub.gap systems. Polyisothianaphthene (PITN) (Wudl et al., 1984), I, and its derivatives (Ikenone et al., 1984), with E.sub.gap .apprxeq.1.1 eV represent some of the more successful experimental realizations of these theoretical predictions (Colaneri, et al., 1986; Kobayashi, et al., 1985). These polymers have E.sub.gap 's 1 eV lower than their corresponding parent, polythiophene, (PT) (Bredas, J. L., 1985). This reduction in E.sub.gap is ascribed (Bredas et al., 1986) to an increased contribution of the quinoid structure, brought about by the 3,4-fused benzene ring. Thus, a considerable amount of the effort to date on narrow band gap polymers has concentrated on increasing their quinoid character. (See FIG. 1).
The energy difference between the aromatic and quinoid structure varies depending on the neutral material's degree of aromaticity. For polymers like polyphenylene, polythiophene, and polypyrrole, it can be substantial so that very little of the quinoid resonance form contributes to the neutral polymer's overall structure. Quinoid segments can be generated in these polymers by the doping process (Bredas, et al., 1984; Bredas, et al., 1982), however, and their growth followed by optical spectroscopy (Chung, et al., 1984). The energy dissimilarity is reduced in PITN since the creation of the quinoid structure in the thiophene moiety is partially compensated by return of aromaticity to the fused six membered ring. This observation has led to several other approaches for generating stable quinoid character. One (Toussaint, et al., 1989) is exemplified by structures like poly(2,7-pyrenylene vinylene), as shown in FIG. 2, structure IIa, to achieve the quinoid resonance form since in doing it exchanges one formally aromatic structure for another (bold outline).
Hence polymer IIa is predicted to have a significantly lower bandgap than the corresponding 1,6 isomer, IIb, (see FIG. 2) which does not have this option (Toussaint, et al., 1989). A second approach does not rely on resonance stabilization to incorporate quinoid character but builds it directly into the monomer and polymer (Toussaint, et al., 1989; Bredas, J. L., 1987; Kertesz, et al., 1987; Hanack, et al., 1991; Jenekhe, S. A., 1986, Wudl, et al., 1988; Zimmer, et al., 1984; Yamamoto, et al., 1981; Miyaura, et al., 1981; Kobmehl, G., 1983). These materials are based on polyarene-methylidenes, III. Neutral films of III (X, Y=S, m=2, n=1) shown in FIG. 3 display absorption maxima around 900 nm (Hanack, et al., 1991), reminiscent of other lowered E.sub.gap polymers like PITN.
Yet another approach to lowered E.sub.gap materials exploits the band crossings between highest occupied (HO) and next highest occupied (NHO) orbitals or lowest unoccupied (LU) and next lowest unoccupied (NLU) orbitals. (Tanaka, et al., 1987; Tanaka, et al., 1988) that occur in certain polymers like polyphenylene and polyperylene.
Theoretically, derivatives with lowered E.sub.gap 's can be obtained by adjusting the frontier orbital occupancy of the polymer. This would be accomplished by replacing certain carbons with either electron rich (e.g., N) or electron poor (e.g., B) elements, their positions carefully chosen, while maintaining planarity. Systems predicted (Tanaka, et al., 1987; Tanaka, et al., 1988) to have lowered E.sub.gap 's are structures IV, V and VI shown in FIG. 4. The synthesis of such materials, however, could be arduous and their processibility is not expected to be high.
When adding heteroatoms, substituents and ring fusions, the symmetry of the frontier orbitals must be considered. Unlike polyacetylene whose bandgap depends primarily on the average bond length alternation (.delta.r), this effect is a secondary contributor to the E.sub.gap of polyheteroaromatics. This parameter is defined as the average of the difference of neighboring long and short C--C bonds. E.sub.gap is a minimum .delta.r=0. (Lowe, et al., 1984; Grant, et al., 1979; Longuet-Higgins, et al., 1959; Kertesz, et al., 1981); Paldus, et al., 1983). The dominant factor for heteroaromatics, however, is the strength of the interaction between the carbon framework and the heteroatom and this is dependent on the symmetry of the former's frontier orbitals (Lee, et al., 1988; Mintmire, et al., 1987). When the highest occupied molecular orbital (HOMO) is antisymmetric and the lowest unoccupied molecular orbital (LUMO) symmetric (as is the case for aromatic arrangements), the band gap increases upon interaction with the heteroatom. The bandgap is decreased, however, for the quinoidal bonding arrangement which has a symmetric HOMO and antisymmetric LUMO (see FIG. 5) (Lee, et al., 1988; Mintmire, et al., 1987). E.sub.gap is minimized at some intermediate structure. Thus polymers such as III, in which the frontier orbitals (HOMO and LUMO) are similarly perturbed by the heteroatom (thus canceling its effect) are expected to have reduced E.sub.gap 's (Lee, et al., 1988; Kertesz, et al., 1989; Lee, et al., 1990).
Polymer VII, formed by annulating a second ring onto PITM, has been predicted by some to be a material with a vanishingly small E.sub.gap. Subsequent calculations (Kertesz et al., 1989; Lee et al., 1990) and experimental measurements (Wudl, et al., 1990) showed that VII (shown in FIG. 6) had a bandgap greater than PITN.